Litcius/Paper detail

Empirical Studies of Institutional Federated Learning For Natural Language Processing

Xinghua Zhu, Jianzong Wang, Zhenhou Hong, Jing Xiao

202051 citationsDOIOpen Access PDF

Abstract

Federated learning has sparkled new interests in the deep learning society to make use of isolated data sources from independent institutes. With the development of novel training tools, we have successfully deployed federated natural language processing networks on GPU-enabled server clusters. This paper demonstrates federated training of a popular NLP model, TextCNN, with applications in sentence intent classification. Furthermore, differential privacy is introduced to protect participants in the training process, in a manageable manner. Distinguished from previous client-level privacy protection schemes, the proposed differentially private federated learning procedure is defined in the dataset sample level, inherent with the applications among institutions instead of individual users. Optimal settings of hyper-parameters for the federated TextCNN model are studied through comprehensive experiments. We also evaluated the performance of federated TextCNN model under imbalanced data load configuration.

Topics & Concepts

Computer scienceFederated learningArtificial intelligenceMachine learningDifferential privacyVariance (accounting)Process (computing)Test dataTraining setSentenceData miningSoftware engineeringAccountingBusinessOperating systemPrivacy-Preserving Technologies in DataCryptography and Data SecurityPrivacy, Security, and Data Protection
Empirical Studies of Institutional Federated Learning For Natural Language Processing | Litcius